Coding Challenge?

Course DS 250

Author

Tyler Bartle

Show the code 
from lets_plot import *
from lets_plot.mapping import as_discrete
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.metrics import accuracy_score

LetsPlot.setup_html(isolated_frame=True)

QUESTION|TASK 1: NAMES

Show the code 
names_df = pd.read_csv("https://raw.githubusercontent.com/byuidatascience/data4names/master/data-raw/names_year/names_year.csv")
peter_df = names_df[(names_df['name'] == 'Peter')]
long_df = peter_df.melt(
    id_vars=['year', 'name'],
    var_name='state',
    value_name='Count'
)
long_df = long_df[long_df['state'].isin(['UT', 'OR'])]
vlines_df = pd.DataFrame({
    'year': [1933, 1969, 2002],
    'label': ['1936', '1972', '2005'],
    'y': [50, 85, 75]
})

x_breaks = list(range(1910, 2021, 5))
x_labels = [str(x) if x % 10 == 0 else '' for x in x_breaks]

p = (ggplot(long_df, aes(x='year', y='Count', color=as_discrete('state')))
 + geom_line(size=1.2)
 + geom_vline(xintercept=1936, color='black')
 + geom_vline(xintercept=1972, color='black')
 + geom_vline(xintercept=2005, color='black')
 + geom_text(data=vlines_df, mapping=aes(x='year', y='y', label='label'), vjust=-1, color='black')
 + ggtitle("The history of Peter for Utah (red) and Oregon (orange)")
 + xlab("Year name given")
 + ylab("Count of Peter")
 + scale_x_continuous(breaks=x_breaks, labels=x_labels)
 + scale_color_manual(values={'UT': 'red', 'OR': 'orange'}, name=None)
 + theme(axis_text_x=element_text(angle=0, hjust=1))
 + ggsize(1200, 500)
)
Show the code 
p.show()

QUESTION|TASK 2: MEAN

Show the code 
problem = pd.Series([np.nan, 15, 22, 45, 31, np.nan, 85, 38, 129, 8000, 21, 2])

median = problem.median()
filled = problem.fillna(median)
mean = round(filled.mean(), 2)

print(mean)
704.75

QUESTION|TASK 3: COUNTRIES

Show the code 
json = [
    { "country":"Afghanistan", "year":1999, "cases":745,    "population":19987071   },
    { "country":"Afghanistan", "year":2000, "cases":2666,   "population":20595360   },
    { "country":"Brazil",      "year":1999, "cases":37737,  "population":172006362  },
    { "country":"Brazil",      "year":2000, "cases":80488,  "population":174504898  },
    { "country":"China",       "year":1999, "cases":212258, "population":1272915272 },
    { "country":"China",       "year":2000, "cases":213766, "population":1280428583 }
]
country_df = pd.DataFrame(json)

# Pivot to get years as columns
pivot_df = country_df.pivot(index='country', columns='year', values='cases').reset_index()
pivot_df.columns.name = None  # Remove MultiIndex on columns

# Rename columns for clarity
pivot_df.columns = ['country', 'cases_1999', 'cases_2000']
pivot_df
country cases_1999 cases_2000
0 Afghanistan 745 2666
1 Brazil 37737 80488
2 China 212258 213766

QUESTION|TASK 4: DWELLINGS

Show the code 
dwellings_df = pd.read_csv("https://raw.githubusercontent.com/byuidatascience/data4dwellings/master/data-raw/dwellings_ml/dwellings_ml.csv")
X = dwellings_df.drop(columns=['parcel', 'basement'])
y = dwellings_df['basement']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20, random_state=2020)
model = GradientBoostingClassifier()
model.fit(X_train, y_train)
y_pred = model.predict(X_test)
acc = accuracy_score(y_test, y_pred)
print(f"Accuracy: {acc:.2%}")
feature_importance = pd.DataFrame({
    'Feature': X.columns,
    'Importance': model.feature_importances_
})

top10 = feature_importance.sort_values(by='Importance', ascending=False).head(10)

p = (ggplot(top10, aes(x='reorder(Feature, Importance)', y='Importance'))
   + geom_bar(stat='identity')
   + coord_flip()
   + ggtitle("Top 10 Feature Importances")
   + xlab("Feature")
   + ylab("Importance")
)
Accuracy: 0.17%
Back to top